Si(110): (16x2) Reconstruction and Adatom Diffusion

Semiconductor surfaces show a wide variety of reconstructions (thechange in atomic positions and bonding due to the surface) which areoften complex and intriguing; they are also enormously important bothfrom scientific and technological viewpoints. Scientifically,understanding the subtle interplay between strain and electronicstructure which give the surfaces many different reconstructions isextremely important; technologically, the growth of structures a fewatomic layers thick is now necessary for the microelectronicsindustry, and will be affected by the starting surface structure. Inthis grant, we will investigate a complex reconstruction on Si(110)with unknown but important structure and also study diffusion ofsilicon and hydrogen atoms across the surface, which will allowcontrolled growth of the surface as well as informing futurestructural studies.The project will be carried out in close collaboration with twoexperimental groups using scanning tunneling microscopy (STM): a teamin Japan for the (16x2) reconstruction; and a group in USA for thehydrogen and silicon diffusion. We will build on a strong trackrecord of experimental interaction and established links with thesegroups to inform and guide our modelling. The project will use highlysophisticated electronic structure methods to model the reconstructionand diffusion, and generate simulated STM images as well as energylevel data and diffusion constants for comparison with otherexperimental techniques.Many of the techniques being developed to extend silicon-basedtransistor technology involve growth or etching of structures onSi(001), e.g. FinFETs or patterned atomic-layer epitaxy (ALE). Thesewill yield structures on Si(001) with Si(110) sides, while otheravenues of research involve fabricating transistors directly onSi(110) for its improved mobility. An ability to understand andcontrol growth on this surface is therefore enormously importanttechnologically, and relies on diffusion of silicon and hydrogen(depending on growth source). We will model the diffusion of bothadatoms on the simple Si(110) surface, along and across the rows inthe substrate and then consider step edges. The final aim of the partwill be to model diffusion around a corner on nano-structure onSi(001), leading to a full understanding of issues affecting growthand overgrowth of Si(110) and structures with Si(110) faces.The reconstructions of Si(001)-(2x1) and Si(111)-(7x7) have alreadybeen solved by a combination of experiment and modelling, though someaspects of Si(001) remain controversial and are the subject ofon-going low temperature experiments; Si(110)-(16x2), however, has notbeen properly understood though some proposals have been made forfeatures on the surface. The structure is fascinating, consisting ofalternately raised and lowered terraces all the same width and runningin the same direction with complex patterns on the terraces; a simplechange to this structure (with all terraces either going up or down atstep edges) leads to the closely related Si(17 15 1) surface. None ofthese aspects have been understood: the terrace width; step edgedirection; the atomic structure on the terraces. We will use carefulelectronic structure modelling in close collaboration with STM andphoto-emission spectroscopy (PES) experiments to propose a surfacestructure, and explain its important features. This will deepen ourunderstanding of semiconductor reconstructions and the mechanismsbehind them, as well as strengthening the UK materials modellingcapability in this area.